One of the challenges facing applications of magnetic nanocrystals is making uniform populations of particles with narrow size and shape distributions. The aim of this proposal is to improve our understanding of how magnetotactic bacteria control the morphology of their crystals. This understanding will allow us to engineer strains to produce nanocrystals superior to synthetic ones. The applications we hope to improve include targeted drug delivery, separation of biological materials from heterogeneous samples, magnetic resonance imaging contrast agents and localized hyperthermia cancer treatments. Magnetotactic bacteria synthesize intracellular, membrane bound crystals with narrow size and shape distributions, and with diverse morphologies. Three publicly available genomes and the development of genetic systems for multiple strains facilitate manipulations of these organisms. This work aims to carry out in vitro characterizations of the matrix proteins on magnetite precipitation and to swap matrix-protein encoding genes from species with different crystal morphologies into more tractable strains. Previous work has demonstrated the ability to select for mutants generated through random and directed evolution, the ability to knock out and replace genes in the magnetotactic bacteria, the initial identification of matrix proteins, the ability to obtain gene sequences from environmental samples of magnetotactic bacteria, and the ability to characterize the magnetic and morphological properties of individual as well as populations of magnetic crystals using ferromagnetic resonance, magnetometry, and transmission electron microscope tomography. This research would further enable the magnetotactic bacteria to be used as a simple model genetic system for understanding biomineralization processes in higher organisms. [unreadable] [unreadable] [unreadable]